By ineke - 9/24/2014
Dear Dave,
I'm working on an experiment in which half of the participants will perform a training and the other half will perform a control task. The two experiments are now programmed separately and ready to use, but we need to get them into one file so that participants can access them online from home (our system only allows us to have one task online). Based on the experiment number, we thought that the even numbers could receive the training and the uneven the placebo, which consist of different trials and thus also out of different blocks. I have tried the 'variables' and 'group assignment' command (see an example below), but all it does is present the two tasks in an other sequence. However, the control group may not receive the block presented in the training group and vice verse. I feel like I'm on the wrong track with these commands, so any tips would be welcome!
Thanks!
<variables>
\groep = (1 of 2) [block1 = intro; block2 = practicetraining; block3= testtraining; block4 = vas]
\groep = (2 of 2) [block1 = intro; block2 = practicecontrol; block3= testcontrol; block4 = vas]
<\variables>
|
By Dave - 9/24/2014
You don't need to merge the experiments into a single script. You can set up a file containing <batch> elements to assing your participants to one of the two tasks. This also works with Inquisit Web (a single license). The script containing the <batch> elements is the one you register as your active web script.
<batch>
/ file = "task_a.iqx"
/ subjects = (1 of 2)
/ groupassignment = groupnumber
</batch>
<batch>
/ file = "task_b.iqx"
/ subjects = (2 of 2)
/ groupassignment = groupnumber
</batch>
|
By ineke - 9/24/2014
Thanks for the quick reply! Meanwhile, I have tried this, but the only thing that happens is a blank screen. Do you have an idea what could have gone wrong?
Thanks
|
By Dave - 9/24/2014
Impossible to answer without the code / launch page.
|
By ineke - 9/24/2014
Well I tried the same as you proposed, but with the correct file names: (I just noticed i posted this wrong, i am working on inquisit 3, could that explain the problem?)
<batch>
/ file = "training.exp"
/ subjects = (1 of 2)
/ groupassignment = groupnumber
</batch>
<batch>
/ file = "control.exp"
/ subjects = (2 of 2)
/ groupassignment = groupnumber
</batch>
|
By Dave - 9/24/2014
groupnumbers don't exist in Inquisit 3, so yes, that's an issue.
With
<batch>
/ file = "training.exp"
/ subjects = (1 of 2)
/ groupassignment = subjectnumber
</batch>
<batch>
/ file = "control.exp"
/ subjects = (2 of 2)
/ groupassignment = subjectnumber
</batch>
what you're doing is assigning *odd* subjecnumbers to "training.exp" and *even* subjectnumbers to "control.exp". You must thus make sure that "training.exp" can deal with *odd* numbers -- as per its <expt> etc. element(s) -- and "control.exp" has things set up to deal with *even* ids. If you have no applicable condition defined, you'll get a blank screen / nothing.
|
By ineke - 10/2/2014
Hi Dave,
Sorry to bother you again, but I have tried different things and I still have had no luck in getting the experiment to run.
When trying to run the batch file, I get the following comment: "File not found. Please verify that the file name path is correct."
However, the file path is correct, the only thing I am not sure of is the <exp> elements that you mentioned in your last post. I could not find it in the tutorials, but how should that look like?
Thanks again!!!
|
By Dave - 10/2/2014
If Inquisit tells you it can't find the file(s), you can be 99.9999% certain that there is an error somewhere in your syntax (a subtle typo, etc.). Check those thoroughly.
|
By ineke - 10/2/2014
That was my first thought too, but the file names are quite simple and are correct. However, currently I am not referring to the batch in the scripts of the separate experiments as I am not clear how to write the <exp> elements, so i assume that is the problem?
|
By Dave - 10/2/2014
No, those are unrelated problems. One step at a time: Attach the files and I can tell you what's wrong.
As for the <expt> elements: You write them as always. The question you need to ask yourself: How many between-subjects conditions are there in your "training" and your "control" script respectively? Then think about which numbers you feed into each via the <batch>. As discussed earlier, "training.exp" only gets odd ids (1,3,5,...), "control.exp" only gets even ids (2,4,6,...). The <expt> elements in those scripts need to account for that.
|
By ineke - 10/2/2014
Ok, thanks! So this is my file that should lead to the 2 different experiments:
<batch >
/ file = "training.exp"
/ subjects = (1 of 2)
/ groupassignment = subjectnumber
</batch>
<batch >
/ file = "control.exp"
/ subjects = (2 of 2)
/ groupassignment = subjectnumber
</batch>
And this is how my current <exp> element looks like in both the separate inquisit files:
<expt>
/ subjects = (1 of 2)
/ blocks = [1 = sessie; 2= practice; 3 = test; 4 = vas]
/ postinstructions = (End)
</expt>
|
By Dave - 10/2/2014
<expt>
/ subjects = (1 of 2)
/ blocks = [1 = sessie; 2= practice; 3 = test; 4 = vas]
/ postinstructions = (End)
</expt>
makes no sense if there is no 2nd <expt> element in the respective script. Delete the /subjects attribute entirely then.
|
By ineke - 10/2/2014
yes that's it!!! Thanks so much an have a nice weekend!
|